地下储存天然气微泄漏地表植被胁迫与适应特征的高光谱探测实验研究

Natural gas is a kind of clean, low-carbon and efficient energy. In order to meet people’s energy needs and ensure secure supplies, the underground natural gas storages are being constructed on a large scale in the central and eastern China. However, there is risk that natural gas may leak from underground, and the traditional methods of detecting the underground natural gas pipeline leakage are not suitable for the underground natural gas storages. However, the natural gas which slightly leaks from underground and enters into soil will lead to stress symptoms of surface vegetation. This project plans to perform controlled simulation experiments in the field and explore an indirectly way of detecting natural gas leakage spots through monitoring changes of vegetations on the sequestration fields by using hyperspectral remote sensing. The research mainly focus on: studying the spread rules and the dynamic three-dimensional spatial distribution of natural gas microseepages in the soil, analyzing the characteristics of surface vegetation stress and adaptation, identifying vegetations under the stress of natural gas leakage and other factors by using hyperspectral remote sensing, extracting the spectral, spatial and time features of vegetation under natural gas leakage stress from hyperspectral remote sensing images and fusing the spectral, spatial and time features together，building identification models of surface vegetation under natural gas leakage stress via fusing the spectral, spatial and time features of the hyperspectral remote sensing images, so as to improve the identification accuracy and reliability of the natural gas leakage spots by using hyperspectral remote sensing. The research can provide the theories and methods for monitoring the underground natural gas storage leakage spots by low-altitude airborne hyperspectral remote sensing in the future, and there are also very important meanings for ensuring the safety of underground natural gas storages.

天然气是一种清洁、低碳、高效的能源，为了满足需求并确保安全供应，我国将在中东部地区大规模建设天然气地下储气库。天然气储存于地下存在泄漏的风险，传统天然气管道探漏方法不适用于储气库，但天然气泄漏进入土壤会导致地表植被出现胁迫症状。本项目拟通过野外控制模拟实验，探索利用高光谱遥感监测地表植被变化而间接探测地下储存天然气微泄漏点的方法。主要研究内容包括：天然气微泄漏在土壤中的传播规律与动态立体空间分布特征，地表植被胁迫与适应特征，高光谱遥感识别天然气微泄漏与其他因素胁迫地表植被的方法，高光谱图像中胁迫植被的光谱、空间及时相特征提取及其特征信息融合方法，构建融合了高光谱图像光谱、空间及时相特征的天然气微泄漏胁迫植被探测模型，以提高高光谱遥感探测天然气微泄漏点的准确度与可靠性。该成果可为未来利用低空机载高光谱遥感探测地下储气库微泄漏点提供理论与方法，对于保证天然气地下储气库安全运营也具有重要意义。

项目针对地下天然气储气库或长输管道微泄漏点难以快速、高效直接探测的问题，围绕利用遥感技术监测地表植被微变化而间接探测天然气微泄漏点的科学目标，通过开展野外定量模拟实验，获取了冠层光谱、近地表高分辨率图像、近地表高光谱成像数据、热红外遥感数据、激光雷达遥感数据、植被生理生化参数、土壤气体浓度、温度与湿度参数，掌握了天然气泄漏进入土壤其浓度空间立体分布规律及其与地表植被变化之间的空间关系，绘制了天然气为泄漏口周围天然气与氧气浓度的立体空间分布图，建立了天然气微泄漏胁迫下植被叶绿素浓度反演模型，构建了基于光谱指数、微分指数、荧光指数、光谱-空间-时相信息融合、热红外遥感等天然气微泄漏胁迫植被识别模型，掌握了天然气微泄漏胁迫植被空间变化特征及植被适应规律，实验结果表明利用多源、多平台数据识别天然气微泄漏点是可行的，且是可信的。这为未来利用高分辨率遥感技术间接探测地下储气库或长输管道微泄漏点提供理论与技术支持。.通过项目组全体人员四年攻关，自主设计并建造了智能化天然气定量微泄漏遥感监测野外试验场，连续开展了4年共7个植被生长季的野外实验，收集了5000余条光谱数据，1TB图像数据，1000余个其他参数，项目资助发表论文20篇，其中SCI 16篇，EI 1篇，中文核心3篇。获发明专利授权2项，出版学术专著1部；项目组2016-2019年共培养研究生7名，本科生毕业论文7名，其中博士生1名，硕士生6名，在读博士生3名，在读硕士生5名。超额完成了任务书规定的各项指标。